Project Summary/ Abstract Acute respiratory tract infections (ARTIs) are a leading cause of antibiotic overuse worldwide. Rapid, accurate identification of a pathogen can reduce antibiotic overuse for ARTIs, but is not always possible with available diagnostics. In low- or middle-income countries (LMICs), ARTI diagnosis is further complicated by limited laboratory infrastructure. Previously, the applicant showed that providing access to rapid antigen-based influenza testing in a LMIC setting was associated with a significant decrease in antibiotic use; however, the majority of influenza-positive patients still received antibiotic prescriptions. Physicians later cited diagnostic uncertainty regarding bacterial infection as a major reason for antibiotic over-prescription. To more effectively minimize antibiotic use for ARTIs, diagnostics that broadly classify infections as bacterial or viral may play an important role. The long-term goal of the applicant is to become an independent investigator developing effective strategies for using novel diagnostics to improve antimicrobial stewardship in LMIC settings. The work proposed in this application will be carried out in a LMIC with the objectives of 1) defining the epidemiology of ARTIs, 2) refining a novel host-based gene expression assay that differentiates between viral and bacterial ARTIs, and 3) determining the performance characteristics of procalcitonin and gene signatures at differentiating between viral and bacterial ARTIs. To accomplish these goals, the applicant will carry out the project under the mentorship of experts in genomics, infectious disease diagnostics, antibiotic resistance/ stewardship, and global health. The proposed Specific Aims will be carried out at a tertiary care hospital in Sri Lanka and include the following: 1) Determine the frequency distribution of viral and bacterial respiratory infections among patients with acute febrile respiratory illness, 2) Refine a gene expression classifier to distinguish viral and bacterial ARTIs, and 3) Prospectively enroll a cohort of ARTI patients to assess the performance characteristics of procalcitonin and gene expression classifiers at differentiating between viral and bacterial ARTIs. By conducting this work, the applicant proposes to fill an important gap in knowledge regarding the applicability of host-based technologies for diagnosing ARTIs in a South Asian, LMIC population. The proposed comprehensive five-year plan includes didactic sessions, hands-on laboratory and bioinformatics work, antimicrobial stewardship and clinical trials training, and structured mentorship to develop the candidate as an independent and successful researcher. The completion of these aims and the acquisition of skills during the training period would place the applicant in a competitive position to apply for future R01 funding. By the conclusion of the award period, the applicant will have contributed significantly to the knowledge of how host-based technologies perform at diagnosing ARTIs in a South Asian, LMIC setting, and will be poised to become a leader in developing effective strategies for improving the rational use of antibiotics in resource-limited settings.